The herpes simplex virus (HSV) thymidine kinase (tk) gene and the gene it overlaps, UL24, provide an excellent model system for studies of eukaryotic gene expression, particularly in HSV-infected cells. HSV and other herpesviruses are important human pathogens. TK mediates antiviral drug sensitivity and drug resistance, which is an increasing problem, especially in patients such as those with AIDS. Thus, these studies are especially health related. This renewal application focuses on novel post-transcriptional events. The first specific aim explores the regulation of UL24 expression whereby a proximal polyadenylation (polyA) signal is utilized early, but not late during infection. The cis-acting sequences responsible will be analyzed by constructing recombinant viruses in which sequences in or near the proximal signal or a more distal signal that is active late in infection or mutant derivatives will be introduced into a gene whose transcription is controlled by a "leaky late" promoter. The roles of HSV ICP27 and US11, which have been implicated in regulation of mRNA 3' end formation, will be examined using virus mutants altered for each of these proteins. The biological role of regulation will be explored by construction of a mutant in which the proximal polyA signal is altered so that if functions late in infection. The second specific aim investigates antisense regulation of UL24 by tk. Whether this regulation requires action in cis will be tested in a coinfection experiment. Potential modifications of UL24 transcripts and their presence in double stranded RNA will be examined by sequencing of cDNA clones and by RNase digestion, respectively. The effects of antisense regulation on UL24 protein synthesis will be assessed using anti-UL24 antisera. The biological role of the antisense regulation will be explored by construcing stable cell lines that overexpress UL24 at early times postinfection. The third specific aim analyzes mechanisms of translational frameshifting of tk mRNA from acyclovir-resistant clinical isolates. To determine if the G-rich frameshifting signal from mutant 615.9 function via Hoogsteen base pairs, a transcript in which the G's are methylated at N7 will be tested for frameshifting. Mutational analyses will be used to correlate the ability of the signal to direct frameshifting and form unusual structures and to test the importance of sequence context. Sequence requirements for frameshifting in yeast will be determined by selection. Other factors important for frameshifting will be identified through suppressor analyses. Other acyclovir-resistant clinical isolates that have caused disease and reactivated from latency will be assayed for frameshifting. These studies link the molecular biology of frameshifting with the clinical problem of drug-resistance.